mikemccand commented on issue #15427: URL: https://github.com/apache/lucene/issues/15427#issuecomment-3672831912
Oooh that sounds great! It would make use of the previous HNSW graphs, whereas today we completely ignore/discard them (except maybe using the largest one as the starting graph for the merged graph)? And it would mean we indeed automatically get that 2nd pass the first time flushed segments are merged? Very cool! Now I really wonder if this would be needle-moving on the recall vs hnsw graph size / visited nodes / search latency! We would use Lucene's segment merging architecture to its advantage, I love it. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
